single-factor-model 0.3.2

Dependencies

• python 3.5

• pandas 0.23.0

• numba 0.38.0

• empyrical 0.5.0

• data_box

• pickle

• multiprocessing

• joblib

Logic

Basic definitions

1. v_t,s_t,c_t: total value, stock value and cash value at time t after trading

2. v^f_t,s^f_t,c^f_t: total value, stock value and cash value at time t before trading

3. ss,sv: suspended stock value and valid stock value

4. r_t: return at time t

5. cost_t: cost to trade at time t

Note: s,ss,sv are all vectors while others are numbers

Equations

1. v_t = |s_t| + c_t

2. s^f_t = s_{t-1} * (1 + r_t) = ss^f_t + sv^f_t = ss_t + sv^f_t

3. ss_t <- suspend, s^f_t

4. c_{t-1} + |sv^f_t| = |sv_t| + c_t + cost_t ( where c_t, cost_t >= 0 )

5. cost_t =|sv_t - sv^f_t| * costRate

6. weight_t <- factor_{t-1},industry_t,suspend_t ( |weight_t| = 1 or 0 if there is no valid stocks or factors or industries)

7. define cost^f_t = (2|sv^f_t| + c_{t-1}) * costRate s.t. cost^f_t >= cost_t, which is greater than the maximum cost we may have during the trade

8. define available_value^f_t = c_{t-1} + |sv^f_t| - cost^f_t, which means the value ( = |sv_t| if weight_t != 0) we have in stocks after trading

9. let sv_t = weight_t * available_value^f_t s.t. c_t = c_{t-1} + |sv^f_t| - |sv_t| - cost_t >=0

Thus to update v_t, we would start with calculating s^f_t, ss, sv^f_t, then cost^f_t, available_value^f_t, then sv_t, cost_t and c_t, and finally v_t

Example

Data Box: pre-process

from data_box import data_box
db=data_box()\
    .load_indestry(ind)\
    .load_indexWeight(ind_weight)\
    .calc_indweight()\
    .load_suspend(sus)\
    .load_adjPrice(price)\
    .add_factor('factor0',factor0)\
    .add_factor('factor1',factor1)\
    .set_lag(freq='d',day_lag=1)\
    .align_data()
# freq can be 'd' or 'm', for detail please refer to db.set_lag doc.

Where price,ind,ind_weight,sus,factor0,factor1 are all dataframes with index as date (yyyymmdd,int) and column as tickers. You can save and load this data box object by db.save('path') and db.load('path'). You can find more in data_box project.

from single_factor_model import run_back_test

Value,Turnover=run_back_test(data_box=db,back_end=None,n=5,out_path=None,double_side_cost=0.003)

Value,Turnover=run_back_test(data_box=db,back_end='loky',n=5,n_jobs=-1,out_path=None,verbose=50)

with __name__=='__main__':
    Value,Turnover=run_back_test(data_box=db,back_end='multiprocessing',n=5,n_jobs=-1,out_path=None)

from single_factor_model import  run_back_test_by_industry
Value_list,Turnover_list=run_back_test_by_industry(db,industry_list=None,back_end='loky',n_jobs=-1,double_side_cost=0.003,verbose=50)

from single_factor_model import calc_return
Return = calc_return(Value,Turnover,long_short=True,double_side_cost=0.003)

from single_factor_model import summary
S=summary(Return)

from single_factor_model import run_plot
run_plot(Return,show=True)